Stick to the SMART values mentioned in the FAQ for failure probability calculation

[saladpanda]

https://www.snapraid.it/faq#smart mentions five SMART values to be of particular importance when judging the disks health:

• Reallocated_Sector_Ct (5)
• Reported_Uncorrect (187)
• Command_Timeout (188)
• Current_Pending_Sector (197)
• Offline_Uncorrectable (198)

However, the calculation performed by snapraid smart also uses Load_Cycle_Count (193), which I find confusing (see #16).

This PR makes snapraid stick to the documented values and avoids confusion.

[amadvance]

Hi @saladpanda

For now, I'm going to put this on hold. I understand that checking this attribute may be somewhat controversial, but on the other hand, I think it's useful to be notified if it increases significantly.

I'll do some research to see if I can find anything conclusive about it.

Ciao, Andrea

[ifsnop]

Can we get the failure probability calculation as a standalone executable or as a bash script? I find it really useful, and would love to use it from inside scripts in different systems (arm/i386). Maybe someone has done some work following this approach.